We have setup a few synchronous relationships which are in sync. On the source volume we do Veeam backups, and for some unknown reason snapshots are "bleeding" over to the destination and left there and we have to clean it up manually... This relationship has been running fine for years, and we think this issue started to happen after upgrade to ontap 9.14.1P5 which we are on now.
Just to be clear it's the default "Synchronous" policy we are using, which has no rules added to it... On the destination we have two snapmirror relation snapshots and x-number of the "VeeamSourceSnapshot....."
And just to make things even more strange, we have 4-5 other relationships setup (same policy) which does not have this issue...
We are of cause considering upgrading to 9.15.1P1 but we would like to know if this issue can be due to mis-configuration (which we doubt)...
The snapshots that are left are nothing special, and they do not have a snapmirror label, but of cause over time they start to fill up the volume 🙂
Oh and we of cause tried to tear down the destination volume and re-create the relationship.. the volumes is about 12TB large and both source and destination are A700 with a 10Gb link for the cluster peering...
#SnapMirror Synchronous "bleeds" snapshots to destination...
tawny perch
minor kernel
I have encountered similar issue with both sync and async mirror snapshots accumulating on Snapmirror Sync Destination volume and not auto deleting. This was with Snapmirror Fan-out with one Source and multiple destinations, one Sync and two async.
There's a diag level option that can be set and documented in KB article
https://mysupport.netapp.com/site/article?lang=en&page=%2Fon-prem%2Fontap%2FDP%2FSnapMirror%2FSnapMirror-KBs%2FSM_Sync_is_not_deleting_SM_created_snapshots_when_Fan-out__with_Mirror_Vault&type=solution
steady cape
sadly, it's only visible to NetApp, not even to Partners
tawny perch
Well sure it is 🙂 I'm a partner and it even has a link to a Burt.. Seems to be fixed in 9.14.1P6 onwards... so another upgrade is in order... I'm looking forward to when NetApp starts to allow minor updates without failovers... The failovers work fine, it's just it takes about 2 hours per HA system to upgrade... 🙂
steady cape
Well sure it is 🙂 I'm a partner and it even has a link to a Burt.. Seems to be...
yeah I see the BURT, but I see no diag command in there anywhere. You must have a different set of permissions than I do 😉
(I mean I'm pretty sure I know what the commands are, but still...)